Skip to content

OpenTelemetry Observability Agent Skills

Universal AgentSkills package providing comprehensive OpenTelemetry guidance for AI coding assistants. Works with Claude Code, GitHub Copilot, Cursor, Windsurf, and other AI-powered development tools.

🚀 Quick Start

Install the skills package:

npx skills add henrikrexed/observability-agent-skills

🎯 What This Provides

A comprehensive collection of specialized skills that help AI assistants write better OpenTelemetry code:

  • Production-Ready Patterns

    Battle-tested patterns from real-world implementations running at scale

  • Security First

    Built-in PII protection, sensitive data redaction, and compliance patterns

  • Kubernetes Native

    Container-optimized configurations and deployment best practices

  • Dynatrace Integration

    Advanced DQL queries, dashboards, SLOs, and automation workflows

📋 Available Skills

Skill Description Best For
otel-instrumentation SDK setup, spans, metrics, logs, PII protection Adding observability to applications
otel-collector Pipeline config, processors, deployment Setting up data collection
otel-ottl Complete OTTL reference and patterns Transforming telemetry data
otel-semantic-conventions Attribute standards, naming patterns Ensuring consistent telemetry
otel-dynatrace DQL, dashboards, SLOs, dtctl CLI Dynatrace-specific workflows

💬 Natural Language Examples

Tell your AI assistant what you want in plain English:

Instrumentation Requests

  • "Add OpenTelemetry tracing to my Express.js API with proper error handling"
  • "Set up Python observability with sensitive data redaction for a Django app"
  • "Configure OpenTelemetry metrics for a Go microservice in Kubernetes"

Collector Configuration

  • "Create an OpenTelemetry collector config for Kubernetes with Dynatrace export"
  • "Write OTTL rules to redact PII from HTTP headers and log messages"
  • "Set up tail-based sampling for high-volume trace data"

Dynatrace Integration

  • "Generate a DQL query to analyze API error rates by service and region"
  • "Create a Dynatrace dashboard for monitoring OpenTelemetry collector health"
  • "Write a workflow to alert on SLO violations with actionable context"

🔧 AI Assistant Compatibility

Claude Code

Skills are automatically available in your workspace. Use with slash commands:

claude instrumentation help  # Get instrumentation guidance
claude collector config     # Generate collector configuration

Cursor

Reference skills in your prompts with @:

@observability Add OTEL tracing to this service
@observability Create collector config for Kubernetes

GitHub Copilot

Skills integrate through comments and chat:

// @observability-agent-skills: setup nodejs instrumentation
// Copilot will use skills to generate appropriate code

VS Code Extensions

Skills work with any MCP/AgentSkills-compatible extension, providing:

  • Contextual code completions
  • Error diagnostics and fixes
  • Intelligent refactoring suggestions

📖 Getting Started

Choose your AI assistant to get specific setup instructions:

  • Claude Code

    Setup for Claude Code with CLAUDE.md and slash commands

  • Cursor

    Configuration with .cursorrules and IDE integration

  • GitHub Copilot

    Setup with copilot-instructions.md for enhanced context

  • Windsurf

    Configuration with .windsurfrules for optimal integration

  • Generic Setup

    Universal setup for any MCP/AgentSkills-compatible assistant

📚 Learning Resources

Created by Henrik Rexed, maintainer of the IsItObservable YouTube channel:

🤝 Contributing

This is an open-source project. Contributions are welcome!

Ready to get started? Choose your AI assistant from the Getting Started section and begin building better observability! 🚀